Working papers

Information transmission in persuasion models with imperfect verification (previously circulated as "Self-Evaluations")(accepted at Theoretical Economics)

Abstract: I study a persuasion game between a privately informed agent and a decision maker (DM) who can imperfectly verify the statements made by the agent by observing a signal that is correlated with the agent's information. I find that whether or not the DM benefits from communicating with the agent depends on whether the DM's signal and the agent's private information satisfy a weak affiliation condition. I then discuss the significance of this result to the debate over the use of self-appraisals in business. I argue that, in general, self-appraisals are only useful when the workers' abilities are multidimensional.

The role of information in collective decisions (joint with Nicolás Figueroa and José-Alberto Guerra) (R&R at the Journal of the European Economic Association)

Abstract:  In this paper, we study how agents who are part of a group that makes decisions using majority rule vote on whether the group should obtain information. We argue, both theoretically and experimentally, that voters are more likely to vote for information to be acquired relative to their own individual willingness to pay for information when ex-ante disagreement is higher and ex-post disagreement is lower. Ex-ante and ex-post disagreement refer to the disagreement among group members over the best policy for the group to follow before and after information is acquired respectively. We discuss the implications of this result on welfare and progress in democratic societies.

Communication through biased intermediators (joint with Nicolás Figueroa and Rubén Jara)

Abstract: We study biased intermediated communication between a sender and a receiver. We prove that the information that can be transmitted from the sender to the receiver is exactly the same as with direct (non-mediated) communication, provided there are (at least) two intermediators who do not communicate with each other. In that sense, the sender is not harmed by not being able to communicate with the receiver directly. This is the case even if the sender does not know the agents' biases and regardless of the rules of communication (cheap talk, bayesian persuasion, etc.). We discuss the implications of our results to a related information design problem, where a decision maker designs the statistical experiment to be performed by a possibly biased agent. We show that, if the decision maker is able to manipulate the data without causing any statistical loss, she can implement her preferred statistical experiment despite the agent being biased. If, however, manipulating the data has a statistical cost, then strategies that reduce the overall informativeness of the experiment (like reducing the sample) might actually be (second-best) optimal for the decision maker, because they may require less costly manipulation in order to dissuade biased agents from misreporting. We discuss these ideas in the context of medical research.

Assignment mechanisms with public preferences and independent types

Abstract: I study a mechanism design problem where a decision maker incentivizes agents to share their private information with her without the use of monetary transfers. I characterize the optimal mechanism under two assumptions: that agents have public preferences (biases) and independent private information. The optimal mechanism resembles a score mechanism except that each alternative's score is also a function of which agents assign points to which alternatives. I apply the model to discuss nepotism and gender discrimination in organizations.